Majid Khonji is an assistant professor in the EECS department at Khalifa University, UAE, leading research activities in the autonomous vehicle laboratory (www.avlab.io) at KUCARS. He is also a research Affiliate with MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), USA. He received his MSc degree in Security, Cryptology, and Coding of Information Systems from Ensimag, Grenoble Institute of Technology, France, and his Ph.D. in Interdisciplinary Engineering from Masdar Institute in collaboration with MIT in 2016.
Previously, he was a visiting assistant professor at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT, a senior R&D technologist at Dubai Electricity and Water Authority (DEWA), and an information security researcher at the Emirates Advanced Investment Group (EAIG).
Dr. Khonji’s research interests include artificial intelligence, robotics, smart grid, theoretical computer science, and optimization.
Risk-Aware Narrow AI for Autonomous Vehicles
In collaboration with Australian National University
A significant barrier to deploying autonomous vehicles (AVs) on a massive scale is safety assurance. Several technical challenges arise from the uncertain environment in which AVs operate, such as road and weather conditions, the uncertain behavior of pedestrians and agent vehicles, and also model inaccuracy. In this proposed work, we will develop an algorithm pipeline along with a technical stack for trajectory optimization for AVs with bounded risk of collision. We consider three major contributions. First, an intention recognition system that predicts the driving-style and the intention of agent vehicles and pedestrians. Second, a planning system that takes into account the uncertainty in the environment and propagates all the way to control policies that explicitly bound the risk of collision. Third, a high-level planning system that extends to multi-vehicle on-the-spot collaboration, under diverse game-theoretic situations that arise due to competing local objectives among AVs. We believe such a white-box approach is crucial for the future adoption of AVs on a large scale.
Development of Autonomous Driving Electric Vehicles based on Infrastructure Sensing
A collaborative research project with KAIST targets infrastructure-based cooperative autonomous driving technology development. In this project, we use infrastructure on the road to compensate for the limitation of the ongoing development of autonomous vehicles. The specific technologies to be developed are 1) vehicle and pedestrian detection from Infracture sensors, 2) infrastructure perception, 3) estimation and mapping of vehicle location, 4) auto parking using magnetic field localization, 5) path following control for an electric vehicle, 6) car-sharing system operation for autonomous vehicles. The final delivery of this project will be an autonomous vehicle driving in an established infrastructure testbed.
Visit www.avlab.io for more updates.